System Engineer

RUNSUN SERVICE PTE. LTD.; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

RUNSUN SERVICE PTE. LTD. in Singapore seeks an experienced AI System Engineer to design, deploy, operate, and optimize AI training clusters and GPU platforms. The role involves managing Linux systems, CUDA/NVIDIA drivers, and container orchestration to ensure high availability.

You will build and maintain scalable AI infrastructure, automate operations, and work with datacenter teams to resolve training environment issues, while supporting on-call rotations and occasional travel as required.

Qualifications

• Bachelor's degree or above in Computer Engineering, Electrical Engineering, Telecommunications, or related fields.
• 3+ years of Linux administration experience; strong knowledge of Ubuntu, Rocky Linux, and RHEL; familiarity with boot process, kernel, filesystems, and performance tuning.
• Ability to troubleshoot complex system issues independently.
• Experience with NVIDIA GPU products H100, H200, B200, B300, GB200 NVL72 and GB300 NVL72.
• Familiar with CUDA, NCCL, NV Link, NV Switch, GPU Direct RDMA.
• Understanding of distributed AI training architectures.
• Hands-on experience with Kubernetes; familiarity with Docker and Containerd; Helm knowledge.

Responsibilities

• Deploy and operate AI training and HPC clusters.
• Install, configure, and optimize operating systems on GPU servers.
• Manage cluster resources and capacity.
• Develop infrastructure automation tools and scripts.
• Build monitoring and observability platforms and ensure high availability.

Skills

Linux administration

GPU computing

Container orchestration

Scripting (Shell, Python)

Network troubleshooting

Education

Bachelor's degree in Computer/Electrical Engineering or related fields

Tools

Kubernetes

Docker

Containerd

NVIDIA CUDA

NCCL

NVIDIA drivers

Job description

We are seeking an experienced AI System Engineer to design, deploy, operate, and optimize AI training clusters, GPU computing platforms, and supporting infrastructure. The ideal candidate should possess strong expertise in Linux systems, GPU computing environments, container platforms, and AI/HPC cluster architectures.

Key Responsibilities

AI Cluster deployment and operation

Deploy and operate AI training and HPC clusters;

Install, configure, and optimize operating systems on GPU servers;

Manage cluster resources and capacity;

Perform system upgrades, patch management, and change implementation;

Develop and maintain standardized operational procedures.

Linux System Management

Manage large-scale Linux environments;

Perform system performance tuning;

Analyze system logs and kernel issues;

Troubleshoot system stability problems;

Manage user access and security policies.

GPU Platform Support

Manage NVIDIA GPU computing platforms;

Deploy and maintain CUDA, NVIDIA Drivers, and Fabric Manager;

Troubleshoot GPU, NV Link, and NV Switch-related issues;

Optimize GPU cluster performance;

Support customer in resolving training environment issues.

Container Platform & Orchestration System

Build and maintain Kubernetes clusters;

Support AI workload scheduling;

Deploy and manage container runtime environments;

Optimize GPU utilization within containers;

Manage Kubernetes high-availability architectures.

AI infrastructure management

Manage distributed storage platforms;

Operate high-speed networking environments;

Collaborate with datacenter teams for troubleshooting;

Monitor infrastructure health and performance;

Improve system reliability and availability.

Automated operation and maintenance and platform development

Develop infrastructure automation tools;

Create deployment and health-check scripts;

Build monitoring and observability platforms;

Implement alerting and self-healing mechanisms;

Improve operational efficiency through automation.

• Good communication, teamwork, and ownership mindset.
  
• Willing to participate in on-call rotation, maintenance windows, and emergency incident response, willing to accept short-term business trips.

Required Qualifications

• Bachelor's degree or above in Computer Engineering, Electrical Engineering, Telecommunications, or related fields.
  
• Linux System
  3+years of Linux administration experience;
  Strong knowledge of Ubuntu, Rocky Linux, and RHEL; Familiarity with system boot process, kernel, filesystems, and performance tuning;
  Ability to troubleshoot complex system issues independently.

• GPU & AI Platform
  Strong understanding of NVIDIA GPU architecture;
  Experience with NVIDIA GPU products H100, H200, B200 B300 ,GB200 NVL72 and GB300 NVL72

• Familiar with CUDA, NCCL, NV Link, NV Switch, GPU Direct RDMA
• Understanding of distributed AI training architectures.

Container and Cloud Native

Hands-on experience with Kubernetes;

Familiarity with Docker and Containerd;

Experience with Helm;

Knowledge of GPU Operator;

Understanding of Kubernetes GPU scheduling.

Networking and Storage

Strong understanding of TCP/IP networking;

Experience with InfiniBand and RoCE;

Familiarity with RDMA architectures;

Experience with one or more storage systems Lustre, BeeGFS, Ceph

and NFS

Automation capabilities

Strong scripting skills in Shell and Python;

Know about Ansible;

Ability to build infrastructure automation scripts.

Preferred Qualities

Experience supporting Large Language Model(LLM) training platforms;

Knowledge of Slurm workload manager;

Experience with Ray and Kubeflow;

Familiarity with NVIDIA Base Command Manager(BCM);

Experience with NVIDIA NIM;

Knowledge of PXE deployment solutions;

Experience on using DDN product;

Experience operating large-scale GPUclusters;

Experience supporting global datacenteroperations.